Search Results (6,019)

Accurately predicting patient clinical outcomes is a complex task that requires integrating diverse factors, including individual characteristics, treatment histories, and environmental influences. This challenge is further exacerbated by missing data and inconsistent data quality, which often hinder the effectiveness of traditional single-task learning (STL) models. Multi-Task Learning (MTL) has emerged as a promising paradigm to address these limitations by jointly modeling related prediction tasks and leveraging shared information. In this study, we proposed MTLNFM, a multi-task learning framework built upon Neural Factorization Machines, to jointly predict patient clinical outcomes on a cohort of 2001 ICU patients. We designed a preprocessing strategy in the framework that transforms missing values into informative representations, mitigating the impact of sparsity and noise in clinical data. We leveraged the shared representation layers, composed of a factorization machine and dense neural layers that can capture high-order feature interactions and facilitate knowledge sharing across tasks for the prediction. We conducted extensive comparative experiments, demonstrating that MTLNFM outperforms STL baselines across all three tasks (i.e., frailty status, hospital length of stay and mortality prediction), achieving AUROC scores of 0.7514, 0.6722, and 0.7754, respectively. A detailed case analysis further revealed that MTLNFM effectively integrates both task-specific and shared representations, resulting in more robust and realistic predictions aligned with actual patient outcome distributions. Overall, our findings suggest that MTLNFM is a promising and practical solution for clinical outcome prediction, particularly in settings with limited or incomplete data, and can support more informed clinical decision-making and resource planning. Full article

This study evaluates the 2022 rollout of the Clean School Bus Rebate Program (CSBRP) to understand how eligibility rules and data practices shape funding distribution across communities with varying needs. We ask whether more accurate maps can improve environmental funding outcomes or whether challenges stem from how agencies define and apply eligibility criteria. Using logistic regression and dasymetric mapping, we find that prioritization criteria helped direct funds to underserved areas, but reliance on school district boundaries introduced inconsistencies that affected program reach. Including charter schools as independent applicants increased competition and sometimes diverted funds from larger public systems serving more. Our geospatial analysis shows that while refined mapping approaches improve resource targeting and reduce goal-outcome mismatches, agency discretion and administrative rules remain key factors in ensuring equitable outcomes. Full article

Petroglyph motifs from 23 sites and 37 panels in northern Georgia and western North Carolina foothills and mountains are analyzed within their archaeological, ethnographic, and landscape contexts. The Track Rock Tradition comprises 10 chronologically sequenced marking categories: (1) Cupules/Meanders/Open Circles; (2) Soapstone Extraction cars; (3) Vulva Shapes; (4) Figures; (5) Feet/Hands/Tracks; (6) Nested Circles; (7) Cross-in-Circles; (8) Spirals; (9) Straight Lines; and (10) Thin Incised Lines. Dating spans approximately 3800 years. Early cupules and meanders predate 3000 years ago, truncated by Late Archaic soapstone extraction. Woodland period (3000–1050 years ago) motifs include vulva shapes, figures, feet, tracks, and hands. Early Mississippian concentric circles date to 1050–600 years ago, while Middle Mississippian cross-in-circles span 600–350 years ago. Late Mississippian spirals (350–200 years ago) and post-contact metal tool incisions represent the most recent phases. The Track Rock Tradition differs from western Trapp and eastern Hagood Mill traditions. Given the spatial overlap with Iroquoian-speaking Cherokee territory, motifs are interpreted through Cherokee beliefs, supplemented by related Muskogean Creek ethnography. In Cherokee cosmology, the matrilocal Thunderers hierarchy includes the Female Sun/Male Moon, Selu (Corn Mother)/Kanati (Lucky Hunter), Medicine Woman/Judaculla (Master of Game), and Little People families. Ritual practitioners served as intermediaries between physical and spirit realms through purification, fasting, body scratching, and rock pecking. Meanders represent trails, rivers, and lightning. Cupules and lines emphasize the turtle appearance of certain rocks. Vulva shapes relate to fertility, while tracks connect to life-giving abilities. Concentric circles denote townhouses; cross-in-circles and spirals represent central fires. The tradition shows continuity in core beliefs despite shifting emphases from hunting (Woodland) to corn cultivation (Mississippian), with petroglyphs serving as necessary waypoints for spiritual supplicants. Full article

This study explores the estimation of the mean of a sensitive variable using calibration estimators under measurement error. Three randomized response techniques are evaluated: Partial Randomized Response Technique, Compulsory Randomized Response Technique, and Optional Randomized Response Technique. Theoretical properties of the proposed estimators are analyzed, and a simulation study using real COVID-19 infection data is conducted. Results indicate that the Optional Randomized Response Technique outperforms Partial Randomized Response Technique and Compulsory Randomized Response Technique in terms of efficiency, underscoring its effectiveness and practical utility for improving data quality in sensitive survey settings. Full article

Background: Quality care of individuals with sickle cell disease (SCD) is dependent upon education of the providers on their care team. Previous studies demonstrate lack of resident and provider comfort regarding care of patients with SCD, yet none have assessed these in medical students. Objective: This study aims to evaluate the adequacy of the research instrument for measuring medical students’ knowledge, confidence, and comfort regarding SCD and related complications prior to wider distribution. Methods: A self-assessment survey was distributed to medical students at two universities to evaluate their knowledge, confidence, and comfort in general SCD topics, in all clinical settings, and regarding common complications. Results: Of the 98 responses, knowledge (p < 0.001) and confidence (p = 0.02) were significantly different between topics, including epidemiology and genetics, pathophysiology, and treatment options. For “treatment options”, there were significant differences in knowledge (p = 0.02) and confidence (p = 0.02) between medical students at different levels of training. Students felt least knowledgeable and least comfortable with care of pregnant women and most knowledgeable and most comfortable with acute pain management. Caring for patients with specific SCD-related conditions increased knowledge and comfort across all domains. Conclusions: This instrument was adequate for measuring knowledge, confidence, and comfort in caring for those with SCD across all clinical settings. We identified a lack of knowledge, confidence, and comfort regarding treatment for those with SCD starting early in medical careers, which improves after caring for patients with various complications. Thus, educating and providing SCD patient experiences is crucial for medical student management confidence related to SCD. Full article

Background: North Carolina Macular Dystrophy (NCMD) is a non-progressive inherited macular dystrophy characterized by marked phenotypic variability. The genetic etiology of NCMD remains largely unknown, and only a limited number of families have been reported in Europe. Methods: We performed an in-depth investigation of an Italian family affected by NCMD using an integrated approach that combined SNP-array analysis, whole-exome sequencing, and long-read whole-genome sequencing. Additionally, we conducted a comprehensive review of NCMD-related literature. Results: We identified a novel 98 Kb duplication involving both PRDM13 and CCNC genes in a three-generation kindred, where the proband exhibited severe macular alterations, while all other affected family members presented with a milder clinical phenotype. A review of the literature suggests different genotype–phenotype correlations and similar penetrance for duplications and single-nucleotide variants (SNVs) in described families. Specifically, smaller duplications may be associated with more severe phenotypes, while SNVs exhibit high phenotypic variability. Conclusions: In this study, we describe the first NCMD Italian family, in which the integration of second- and third-generation sequencing methods enabled the identification of a novel pathogenic PRDM13 and CCNC duplication, thereby expanding the mutational spectrum of NCMD. Overall, these findings, together with the literature review, highlight the importance of selecting appropriate genetic testing approaches that allow the detection of non-coding variants and CNVs and thus enable accurate diagnosis and effective clinical management of patients and their families. Full article

In the context of typhlo music therapy, personalized interventions can significantly enhance the therapeutic experience for visually impaired children. Leveraging a data-driven approach, we incorporate action-rule discovery to provide insights into the factors of music that may benefit individual children. The system utilizes a comprehensive dataset developed in collaboration with an experienced music therapist, special educator, and clinical psychologist, encompassing meta-decision attributes, decision attributes, and musical features such as tempo, rhythm, and pitch. By extracting and analyzing these features, our methodology identifies key factors that influence therapeutic outcomes. Some themes discovered through action-rule discovery include the effect of harmonic richness and loudness on expression and communication. The main findings demonstrate the system’s ability to offer personalized, impactful, and actionable insights, leading to improved therapeutic experiences for children undergoing typhlo music therapy. Our conclusions highlight the system’s potential to transform music therapy by providing therapists with precise and effective tools to support their patients’ developmental progress. This work shows the significance of integrating advanced data analysis techniques in therapeutic settings, paving the way for future enhancements in personalized music therapy interventions. Full article

In recent years, online streaming has encountered the challenge of retaining its user base. This study considers the role of transaction cost economics theory in consumer choices to continue subscribing. Participants respond to their top three streaming services, resulting in 797 responses, accounting for multiple selections by each respondent. Respondents could choose their top three services from a list of Netflix, Disney, Hulu, Amazon Prime Video, HBO Max, and Apple TV+. The study’s conclusions highlight the impact of uncertainty, a negative measure of streaming quality, on online subscription-based video streaming. Additionally, asset specificity, reflecting uniqueness and exclusive content, is found to be positively related to continuing a subscription. This research distinguishes itself by examining individuals who are already subscribers to provide insights and guidance through the lens of Transaction Cost Economics, to help marketing professionals seeking a deeper understanding of consumer behavior in the online streaming landscape. Full article

Ovarian cancer is a deadly gynecological malignancy that will affect about 21,000 women and result in almost 153,000 deaths in the United States in 2025. New clinical tools that facilitate early diagnosis and treatment of ovarian malignancies will significantly help reduce mortality and improve current long-term survival rates. We utilized a previously identified single-strand DNA aptamer RLA01 that binds and internalizes into target epithelial ovarian cancer cells to label PLGA-based nanoparticles and determine their ability to selectively target EOC cells and deliver payloads for cellular internalization. Nanoparticles labeled with RLA01 significantly enhanced cellular uptake 20–85% by receptor-mediated endocytosis into target EOC Caov-3 cells and inhibited cellular uptake in non-target HOSE 6-3 cells. Further, labeling of paclitaxel-loaded nanoparticles with RLA01 significantly decreased cell proliferation and induced apoptosis. A preliminary pilot study looking at the in vivo stability of aptamers demonstrated their ability to promote retention and honing of nanoparticles at tumors. These data demonstrate the effective combinatorial use of aptamer RLA01 and nanoparticle technologies for the direct targeting of tumor cell populations both in vitro and in vivo. Full article

We consider the problem of privately updating a message out of K messages from N replicated and non-colluding databases where a user has an outdated version of the message ${\hat{W}}_{\theta }$ of length L bits that differ from the current version $W_{\theta }$ in at most f bits. The user also has a cache containing coded combinations of the K messages (with a pre-specified structure), which are unknown to the N databases (unknown prefetching). The cache Z contains ℓ linear combinations from all K messages in the databases with $r=\frac{\mathcal{l}}{L}$ being the caching ratio. The user needs to retrieve $W_{\theta }$ correctly using a private information retrieval (PIR) scheme without leaking information about the message index $\theta$ to any individual database. Our objective is to jointly design the prefetching (i.e., the structure of said linear combinations) and the PIR strategies to achieve the least download cost. We propose a novel achievable scheme based on syndrome decoding where the cached linear combinations in Z are designed to be bits pertaining to the syndrome of $W_{\theta }$ according to a specific linear block code. We derive a general lower bound on the optimal download cost for $0\le r\le 1$, in addition to achievable upper bounds. The upper and lower bounds match for the cases when r is exceptionally low or high, or when $K=3$ messages for arbitrary r. Such bounds are derived by developing novel cache-aided arbitrary message length PIR schemes. Our results show a significant reduction in the download cost if $f<{\displaystyle \frac{L}{2}}$ when compared with downloading $W_{\theta }$ directly using typical cached-aided PIR approaches. Full article

Intratumoral injections of macromolecules, such as biologics and immunotherapeutics, show promise in overcoming dose-limiting side effects associated with systemic injections and improve treatment efficacy. However, the retention of injectates in the tumor microenvironment is a major underappreciated challenge. High interstitial pressures and dense tumor architectures create shear forces that rapidly expel low-viscosity solutions post-injection. Injectable hydrogels may address these concerns by providing a viscoelastic delivery vehicle that shields loaded therapies from rapid expulsion from the tumor. A chitosan–glycerol hydrogel was thus developed and characterized with the goal of improving the injection retention of loaded therapeutics. The gelation parameters and mechanical properties of the hydrogel were explored to reveal a shear-thinning gel that is injectable through a 27-gauge needle. Biocompatibility studies demonstrated that the chitosan–glycerol hydrogel was nontoxic. Retention studies revealed significant improvements in the retention of model therapeutics when formulated with the chitosan–glycerol hydrogel compared to less-viscous solutions. Finally, release studies showed that there was a sustained release of model therapeutics of various molecular sizes from the hydrogel. Overall, the chitosan–glycerol hydrogel demonstrated injectability, enhanced retention, biocompatibility, and sustained release of macromolecules, indicating its potential for future clinical use in intratumoral macromolecule delivery. Full article

A comprehensive metabolic profiling of Catharanthus roseus (L.) G. Don was performed using tandem mass spectrometry, along with an evaluation of the biological activities of its various solvent extracts. Among these, the methanolic leaf extract exhibited mild radical scavenging activity, low to moderate antimicrobial activity, and limited cytotoxicity in both the brine shrimp lethality assay and MTT assay against HeLa and A549 cell lines. High-performance liquid chromatography–electrospray ionization–high-resolution tandem mass spectrometry (HPLC-ESI-HRMS/MS) analysis led to the annotation of 34 metabolites, primarily alkaloids. These included 23 indole alkaloids, two fatty acids, two pentacyclic triterpenoids, one amino acid, four porphyrin derivatives, one glyceride, and one chlorin derivative. Notably, two metabolites—2,3-dihydroxypropyl 9,12,15-octadecatrienoate and (10S)-hydroxypheophorbide A—were identified for the first time in C. roseus. Furthermore, Global Natural Products Social Molecular Networking (GNPS) analysis revealed 18 additional metabolites, including epoxypheophorbide A, 11,12-dehydroursolic acid lactone, and 20-isocatharanthine. These findings highlight the diverse secondary metabolite profile of C. roseus and support its potential as a source of bioactive compounds for therapeutic development. Full article

Understanding the interplay between lightning activity and hydrometeor distribution is crucial for advancing knowledge of thunderstorm electrification processes. Using three-dimensional lightning mapping and dual-polarization radar observations, this study investigates the spatiotemporal correlations between low-frequency (LF) lightning sources and hydrometeors during a severe thunderstorm on 11 June 2014, in North Carolina, USA. The results reveal that lightning sources are predominantly observed above 6 km (near the −10 °C isotherm) and stabilize into a dual-peak vertical distribution as the storm progresses into its mature stage, with peaks located at 6–7 km (−10 °C to −15 °C) and 10–11 km (approximately −40 °C). Low-density graupel (LDG) and aggregates (AGs) dominate at lightning locations. Stronger updrafts lead to higher proportions of LDG and high-density graupel (HDG), and lower proportions of AG. LDG exhibits the strongest positive correlation with LF lightning sources, with a peak correlation coefficient of 0.65 at 9 km. During the vigorous development stage, HDG and hail (Ha) also show positive correlations with LF lightning sources, with peak correlation coefficients of 0.52 at 7 km and 0.42 at 8 km, respectively. As the storm reaches its mature phase, the correlation between LDG and lightning sources also displays a dual-peak vertical distribution, with peaks at 7–8 km and 13–14 km. Both the peak correlation coefficient and its corresponding height increase with the strengthening of updrafts, underscoring the critical role of updrafts in microphysical characteristics and driving electrification processes. Full article

The Mid-Piacenzian Warm Interval (MPWI) is marked by warmer temperatures and higher atmospheric CO₂ levels than today, making it an analogue for late-21st-century-warming, whereas the early Pleistocene cooling is more like today. We compare seasonal growth temperatures derived from oxygen isotope ratios (δ¹⁸O) and clumped isotopes (∆₄₇) in Mercenaria. Modern shells were previously collected from coastal NC. The fossil shells are from the Duplin (MPWI) and Waccamaw Formations (early Pleistocene), NC. Oxygen isotope ratios range from −2.2‰ to 2.3‰ (modern), −0.9‰ to 2.4‰ (MPWI), and −0.9‰ to 2.9‰ (early Pleistocene). The values of Δ₄₇ range from 0.576‰ to 0.639‰ (modern), 0.566‰ to 0.621‰ (MPWI), and 0.581‰ to 0.615‰ (early Pleistocene). We show that Mercenaria do not require a species-specific ∆₄₇ calibration. Modern and MPWI ∆₄₇-derived summer/winter temperatures (SST_∆47) and seasonal amplitudes are indistinguishable from δ¹⁸O-derived temperatures. The early Pleistocene summer SST_∆47 is indistinguishable from δ¹⁸O-derived temperatures, but the winter SST_∆47 is warmer by 5 °C and may reflect within-shell time averaging. The modern summer/winter SST_∆47 are indistinguishable from the MPWI, but the MPWI has a lower seasonal amplitude by 5 °C. Compared to our calculated δ¹⁸O_sw values, modeled values for the MPWI are within error but are much lower, and they are not within error for the early Pleistocene. Full article

Prolonged inundations are altering coastal forest ecosystems of the southeastern US, causing extensive tree die-offs and the development of ghost forests. This hydrological stressor also alters carbon fluxes, threatening the stability of coastal carbon sinks. This study was conducted to investigate the interactions between hydrological drivers and ecosystem responses by analyzing daily eddy covariance flux data from a wetland forest in North Carolina, USA, spanning 2009–2019. We analyzed temporal patterns of net ecosystem exchange (NEE), gross primary productivity (GPP), and ecosystem respiration (RE) under both flooded and non-flooded conditions and evaluated their relationships with observed tree mortality. Generalized Additive Modeling (GAM) revealed that groundwater table depth (GWT), leaf area index (LAI), NEE, and net radiation (Rn) were key predictors of mortality transitions (R² = 0.98). Elevated GWT induces root anoxia; declining LAI reduces productivity; elevated NEE signals physiological breakdown; and higher Rn may amplify evapotranspiration stress. Receiver Operating Characteristic (ROC) analysis revealed critical early warning thresholds for tree mortality: GWT = 2.23 cm, LAI = 2.99, NEE = 1.27 g C m⁻² d⁻¹, and Rn = 167.54 W m⁻². These values offer a basis for forecasting forest mortality risk and guiding early warning systems. Our findings highlight the dominant role of hydrological variability in ecosystem degradation and offer a threshold-based framework for early detection of mortality risks. This approach provides insights into managing coastal forest resilience amid accelerating sea level rise. Full article